Predictive Factors For Postoperative Complications Research Articles

Predictors of postoperative complications and survival in patients with major salivary glands malignancies: A study highlighting the influence of age

The purpose of this study was to reveal prognostic factors for surgical complications and survival in young and elderly patients with malignant salivary gland tumors. Retrospective analysis of surgically treated patients with malignant major salivary gland tumors was performed. Of the 111 patients, 33 patients were 70 years or older and 78 patients were under 70. Comorbidity was recorded using the Adult Comorbidity Evaluation-27 index and complications using the Clavien-Dindo classification. Comorbidity and complications were significantly more frequent in elderly patients (p < .05). Age, comorbidity, and length of surgery were not significant independent predictors of complications. Stage was the only independent predictive factor for postoperative complications and disease-specific survival. Based on this retrospective analysis reviewing surgically treated patients, age alone should not be a reason to treat elderly patients differently, as it was not a predictor of either complications or disease-specific survival.

Pulmonary resections in a tertiary care center—a prospective observational study of outcome

Introduction In order to assess short term outcomes of pulmonary resections in one year, a prospective observational study was made. To assess the predictive factors for postoperative complications.

Risk Factors for Perioperative Complications of Laparoscopic Adrenalectomy Including Single-Site Surgery

To determine predictive factors for perioperative complications of laparoscopic adrenalectomy (LA) including single-site surgery performed or supervised by experienced laparoscopic surgeons. We retrospectively reviewed 265 consecutive unilateral LAs performed at a single institution from 2001 to 2011. All surgical procedures were performed or supervised by one of two experienced laparoscopic surgeons. The transabdominal approach was used in 248 patients, while a retroperitoneal approach was used in 17 patients. Laparoendoscopic single-site adrenalectomy was performed in 58 patients. Predictive factors for postoperative complications were analyzed. There were no postoperative deaths. At least one postoperative complication occurred in 12 (4.6%) patients. No patient underwent reoperation for a complication. Only one (0.4%) patient's procedure was converted to open surgery, and only this patient needed a blood transfusion during the operation. One patient who underwent laparoendoscopic single-site adrenalectomy had the procedure converted to conventional laparoscopic adrenalectomy. Multivariable logistic regression analysis showed that an American Society of Anesthesiologists-physical status grade 3 or above was the only predictive factor for perioperative complications. American Society of Anesthesiologists-physical status grade 3 or above was a statistically significant factor for perioperative complications associated with laparoscopic adrenalectomy that involved laparoendoscopic single-site adrenalectomy, which was performed or supervised by an experienced laparoscopic surgeon. According to the findings of this retrospective study, single-site operation, obesity, adrenal tumor size, and tumor type may not be considered as risk factors for perioperative complications after LA performed or supervised by an experienced laparoscopic surgeon.

Importance of recurrence rating, morphology, hernial gap size, and risk factors in ventral and incisional hernia classification

There is limited evidence on the natural course of ventral and incisional hernias and the results of hernia repair, what might partially be explained by the lack of an accepted classification system. The aim of the present study is to investigate the association of the criteria included in the Wuerzburg classification system of ventral and incisional hernias with postoperative complications and long-term recurrence. In a retrospective cohort study, the data on 330 consecutive patients who underwent surgery to repair ventral and incisional hernias were analyzed. The following four classification criteria were applied: (a) recurrence rating (ventral, incisional or incisional recurrent); (b) morphology (location); (c) size of the hernial gap; and (d) risk factors. The primary endpoint was the occurrence of a recurrence during follow-up. Secondary endpoints were incidence of postoperative complications. Independent association between classification criteria, type of surgical procedures and postoperative complications was calculated by multivariate logistic regression analysis and between classification criteria, type of surgical procedures and risk of long-term recurrence by Cox regression analysis. Follow-up lasted a mean 47.7 ± 23.53 months (median 45 months) or 3.9 ± 1.96 years. The criterion "recurrence rating" was found as predictive factor for postoperative complications in the multivariate analysis (OR 2.04; 95 % CI 1.09-3.84; incisional vs. ventral hernia). The criterion "morphology" had influence neither on the incidence of the critical event "recurrence during follow-up" nor on the incidence of postoperative complications. Hernial gap "width" predicted postoperative complications in the multivariate analysis (OR 1.98; 95 % CI 1.19-3.29; ≤5 vs. >5 cm). Length of the hernial gap was found to be an independent prognostic factor for the critical event "recurrence during follow-up" (HR 2.05; 95 % CI 1.25-3.37; ≤5 vs. >5 cm). The presence of 3 or more risk factors was a consistent predictor for "recurrence during follow-up" (HR 2.25; 95 % CI 1.28-9.92). Mesh repair was an independent protective factor for "recurrence during follow-up" compared to suture (HR 0.53; 95 % CI 0.32-0.86). The ventral and incisional hernia classification of Dietz et al. employs a clinically proven terminology and has an open classification structure. Hernial gap size and the number of risk factors are independent predictors for "recurrence during follow-up", whereas recurrence rating and hernial gap size correlated significantly with the incidence of postoperative complications. We propose the application of these criteria for future clinical research, as larger patient numbers will be needed to refine the results.

Impact of comorbidity on complications after nephrectomy: use of the Clavien Classification of Surgical Complications

What's known on the subject? and What does the study add? Tumour characteristics, physical status and comorbidities are considered important for surgical outcome and prognosis. The present study objectively evaluates the association between comorbidity and postoperative complications after nephrectomy for RCC, by using the modified Clavien Classification of Surgical Complications to grade complications after nephrectomy. To present a single-centre experience of open nephrectomy for lesions suspected for renal cell carcinoma (RCC), evaluating the association between comorbidity and postoperative complications using a standardized classification system for postoperative complications. Clinicopathological data of 198 patients undergoing open radical or partial nephrectomy for lesions suspected of RCC were retrospectively analysed. Comorbidity scored by the Charlson comorbidity index (CCI), body mass index, age, gender, surgical procedure and surgical history were examined as predictive factors for postoperative complications, which were scored using the modified Clavien Classification of Surgical Complications (CCSC). The overall complication rate was 34%: 7% grade I, 15% grade II, 5% grade III, 3% grade IV and 4% grade V. Preoperative comorbidities were present in 51% of all patients. There were significantly more major complications (CCSC >2) in patients with major comorbidities (CCI >2), at 16% vs 7% (P = 0.018). Patients with high-stage RCC had significantly more severe complications than low-stage RCC (P = 0.018). In multivariable analysis, comorbidity (odds ratio [OR] 7.55, P = 0.004) and tumour stage 3-4 (OR 6.23, P = 0.007) were independent predictive factors for major complications. Major complications occur significantly more often when major comorbidities are present. Comorbidity scores can be used in risk stratification for complications and should be considered during decision-making and counselling of patients before nephrectomy.

Risk factors for conversion and complications after unilateral laparoscopic adrenalectomy

Laparoscopic adrenalectomy (LA) is the procedure of choice for surgical management of most benign adrenal tumours, with a reported overall complication rate around 10 per cent. The aim of this study was to determine predictive factors for postoperative complications and conversion to open surgery after unilateral LA. From 1994 to 2009, consecutive patients undergoing unilateral LA by the lateral transabdominal approach were analysed from a prospectively maintained database. A mass larger than 12 cm in diameter and suspected primary adrenal carcinoma were considered contraindications to LA. Predictive factors for postoperative complications and conversion to open surgery were analysed. Some 462 patients were analysed. There were no postoperative deaths. Postoperative complications occurred in 53 patients (11·5 per cent), medical complications in 28, and surgical complications in 33 patients. Six patients underwent reoperation for complications. Multivariable logistic regression analysis showed that conversion to open surgery (odds ratio (OR) 6·20, 95 per cent confidence interval 2·08 to 18·53; P = 0·001) and left-sided tumour (OR 1·89, 1·02 to 3·52; P = 0·044) were independent predictive factors for overall complications. Conversion to open surgery was the only independent predictive factor for medical complications (OR 12·88, 4·21 to 39·41; P = 0·001), and left-sided LA was the only predictive factor for surgical complications (OR 2·22, 1·01 to 4·89; P = 0·047). No factor was predictive of conversion to open surgery. In this single-institution study, conversion to open surgery and left-sided tumours were independent predictive factors for overall complications, but none of the variables analysed was predictive of conversion.

Pulmonary Gas Exchange and Plasma Lactate in Horses with Gastrointestinal Disease Undergoing Emergency Exploratory Laparotomy: A Comparison with an Elective Surgery Horse Population

To characterize pulmonary gas exchange and arterial lactate in horses with gastrointestinal disease undergoing anesthesia, compared with elective surgical horses, and to correlate these variables with postoperative complications and mortality. Prospective clinical study. Horses undergoing emergency laparotomy for acute intestinal disease (n = 50) and healthy horses undergoing elective surgery in dorsal recumbency (n = 20). Arterial blood gas analysis was performed at predetermined intervals on horses undergoing a standardized anesthetic protocol. Alveolar-arterial oxygen gradient was calculated. Predictive factors for postoperative complications and death in colic horses were determined. Arterial oxygen tension (P(a) O(2)) varied widely among horses in both groups. P(a) O(2) significantly increased in the colic group after exteriorization of the ascending colon. P(a) O(2) and alveolar-arterial oxygen gradient were not significantly different between groups, and neither were correlated with horse outcome. Arterial lactate in recovery ≥ 5 mmol/L was associated with a 2.25 times greater relative risk of complications and lactate ≥ 7 mmol/L was associated with a 10.5 times higher relative risk of death. Colic horses in this population were not more likely to be hypoxemic than elective horses, nor was gas exchange impaired to a greater degree in colic horses relative to controls. Arterial lactate sampled immediately after anesthetic recovery was predictive for postoperative complications and death.

Factors Affecting Morbidity/Mortality of Emergency General Surgery

Introduction/Background: Emergency status has been documented as adversely affecting the outcome of general surgery. However, the predictors of increased morbidity and mortality of emergency general surgery (EGS) are unclear, including the possibility of modifying these factors. This study aims to determine the predictive factors for postoperative complications of EGS. Methods: A retrospective study of all randomly selected Brigham and Women's Hospital (BWH) American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) patients who had an EGS procedure from January 1, 2007 to December 31, 2008 were selected for this study. BWH IRB approved this study. The study applied the inclusion and exclusion criteria as defined by ACS NSQIP using select preoperative and all postoperative data variables. Additional non-ACS NSQIP data points were added and analyzed as possible predictors of poor outcomes. These included blood glucose levels closest to the time of surgical incision, length of time from ED arrival to incision, total volume of intravenous fluids (IVF) administered within 24 hours prior to incision, and time of antibiotic administration prior to incision. Our primary outcomes were whether or not the patient had experienced any of the pre-defined ACS NSQIP postoperative occurrences within 30 days and/or died while hospitalized for the target emergency procedure. Univariate statistical analyses were done using Wilcoxon signed rank sum test for continuous variables and Chi square test/Fisher Exact test for categorical variables. Multivariate analyses were done to identify independent predictors using logistic regression. Results: We identified 263 cases. 46.4% were male. 60 (22.8%) had complications, and 25 (9.5%) died within 30 days. Those without complications had a mean age of 51 years while those with complications had a mean age of 66 years. The most common procedures were appendectomies (22.4%), colectomies (18.6%) and enterectomies (17.1%). The most common postoperative occurrences were prolonged ventilator support>48hrs (12%), unplanned intubations (6%), and superficial incisional infections (5%). In univariate analyses, age (p<0.0001), male gender (p<0.001), length of operation (p<0.0001), preoperative blood transfusion (p<0.001), elevated blood glucose (p=0.026), elevated BUN (p<0.0001), functional health status (p=0.007), ASA (p<0.001), presence of COPD (p=0.002), and history of smoking (p=0.009) were found to be significant predictors of postoperative complications. Length of time from ED arrival to time of incision (p=0.42), total volume of IVF administered within 24 hours prior to incision (p=0.19), and time of antibiotic administration prior to incision (p=0.74) were not found to be significant. Multivariate analyses using stepwise logistic regression found male gender (OR 2.01, 95% CI 1.05, 4.10), smoking (OR 2.69, 1.15, 6.26), elevated BUN (OR 1.04, 95% CI 1.02, 1.06), and length of operation time (OR 1.001, 95% CI 1.000, 1.003) to be independent predictors for postoperative complications. Conclusions: Emergency general surgery patients who develop postoperative complications are more likely to be male, smokers, have a higher preoperative BUN, and experience longer operative times. Preoperative fluid resuscitation to address elevated BUN and experienced surgical teams to decrease operative times are potential targets to improve outcomes in emergency general surgery patients.

Surgical Lung Resection for Severe Hemoptysis

The role of surgical lung resection in the management of severe hemoptysis has evolved after advances in interventional radiology. We sought to describe the indications for surgical lung resection in such patients and to identify predictive factors of postoperative complications. This study is a retrospective analysis (May 1995 to July 2006) of consecutive patients referred to the intensive care unit of a tertiary hospital for severe hemoptysis who underwent surgical lung resection. Among 813 patients referred for severe hemoptysis, 111 underwent surgical lung resection. Interventional radiology had been first attempted in 87 patients (78%); 68 underwent surgery because of a failed procedure (n = 28) or bleeding persistence or recurrence within 72 hours despite a completed procedure (n = 40); 19 patients underwent surgery after bleeding control. The remaining 24 patients (22%) were directly referred to the surgeon (5 for emergency surgery). Overall, surgery was performed in emergency (n = 48), scheduled after bleeding control (n = 48), or planned after discharge (n = 15). The main indications for surgery were mycetoma, cancer, bronchiectasis, and active tuberculosis. Surgery for mycetoma (odds ratio, 9.4; 95% confidence interval, 2.8 to 32), emergency surgery (odds ratio, 5.3; 95% confidence interval, 1.8 to 16), and pneumonectomy (odds ratio, 4.7; 95% confidence interval, 1.2 to 18) independently predicted complications. Fifteen patients died in the intensive care unit, of whom 14 underwent emergency surgery. Chronic alcoholism (odds ratio, 4.6; 95% confidence interval, 1.1 to 19), the need for mechanical ventilation or vasoactive drugs on admission (odds ratio, 8.2; 95% confidence interval, 1.9 to 35), and blood transfusion before surgery (odds ratio, 8; 95% confidence interval, 1.5 to 42) predicted mortality. Attempting at controlling bleeding with first-line nonsurgical approaches appears necessary to optimize the operative conditions and improve outcome of patients with severe hemoptysis.

Patients with Less Than Three Episodes of Diverticulitis May Benefit from Elective Laparoscopic Sigmoidectomy

This study was designed to asses the predictive factors of postoperative complications in patients who underwent a laparoscopic elective approach for recurrent diverticulitis and to determine the relationship between the number of acute episodes and surgical morbidity. A retrospective analysis was performed on patients with colonic diverticular disease treated by an elective laparoscopic approach between July 2000 and November 2007. The variables studied were age, sex, BMI, ASA, number of previous acute episodes, local severity, abdominal surgery history, comorbidity, and laparoscopic training of the surgeon. Logistic regression analysis was used to establish significant results. A total of 137 patients were analyzed; 87 (63.5%) were men with a mean age of 56.7 (range, 27-89) years. Intraoperative and postoperative complications occurred in 2.9% (n = 4) and 12.4% (n = 17) of the patients respectively. Conversion rate was 9.4% (n = 13). Local severity (odds ratio (OR), 16.34; 95% confidence interval (CI), 4.1-64.5, p = 0.00007), history of abdominal surgery (OR, 3.02; 95% CI, 0.8-11.5; p = 0.02), and the training of the operating surgeon (OR, 4.8; 95% CI, 1.02-22.7; p = 0.001) were significant risk factors related to surgery conversion. A history of three or more acute episodes was significantly associated with a high severity of local process and was a risk factor related to conversion (OR, 2.6; 95% CI, 0.5-12.3; p = 0.22). The severity of the local process seems to be a risk factor for perioperative complications. A significant association (chi2, 4.45; p = 0.03) between conversion and postoperative complications also was observed (OR: 3.79, 95% CI, 1.02-14.07; p = 0.04). A history of three or more acute episodes of diverticulitis with conservative treatment is associated with a high severity of the local process during laparoscopic sigmoidectomy and increases the rate of conversion and perioperative complications.

Surgery for body packing in the Caribbean: a retrospective study of 70 patients.

Body packing is a way to deliver packages of drugs hidden in body cavities, across international borders. This may cause mechanical bowel obstruction and intoxication leading to emergency surgery. The objective of this retrospective study is to evaluate pre-, per- and postoperative characteristics of body packing and its complications and prognosis after surgery. Medical information was obtained from 70 patients diagnosed with Body Packer syndrome who underwent surgery for body packing in the period of July 2001-July 2005 at the St. Elisabeth hospital, Curaçao, the Netherlands Antilles. Wound infection occurred in 32.9% and fascia dehiscence in 2.9%. Most procedures were enterotomies. One patient had a re-laparotomy for Hartmann's procedure because of an anastomotic leak and three other patients had a partial small bowel resection. The creation of a stoma was only needed in one patient. One patient died of postoperative DIC. In six patients packages were left behind, not requiring re-operation. Bowel obstruction and symptoms of intoxication are clear indications for surgery in body packing patients. In our study there was a low mortality and stoma placement rate, we did find a remarkable high amount of wound infection but no evidence for a higher incidence of fascia dehiscence in comparison with elective gastrointestinal surgery. In surgery for bodypacking a mandatory postoperative x-ray is indicated. A prospective trial is needed to assess more information about possible predictive factors of postoperative complications and mortality after surgery for body packing.

Predictive factors for complications in elderly patients who underwent head and neck oncologic surgery

Postoperative complications are relevant outcomes in patients with head and neck tumor who have undergone surgery. Few trials have assessed predictive factors in older patients. We assessed the predictive effect of preoperative clinical factors on postoperative complications. We conducted a cohort study with 242 patients older than 70 years with head and neck cancer who underwent surgery. Logistic regression identified predictive factors for postoperative complications. Significant variables were used to build a predictive index. Comorbidities were present in 87.6% of patients, and 56.6% had some type of complication (44.6% local and 28.5% systemic). Male sex, bilateral neck dissection, presence of 2 or more comorbidities, reconstruction, and clinical stage IV were associated with postoperative complications. The predictive index showed a receiver operating characteristics curve (ROC) area of 0.69. It is possible to predict postoperative complications in older patients with head and neck tumors who underwent oncologic surgery using clinical preoperative variables.

Early Postoperative Response of Cytokines in Liver Transplant Recipients

We evaluated the early postoperative response of several cytokines (IL-2, IL-4, IL-6, IL-10, TNF-α, IFN-γ) prior to liver transplantation (T 0) as well as 1, 6, and 12 hours and 1, 2, 3, 5, and 7 days afterward. Cytokine concentrations were correlated with serum levels of bilirubin as a predictor of postoperative complications. Cytokine levels were determined in plasma samples from 16 liver transplant recipients (13 men, 3 women) aged 43 to 61 years. IL-6 and IL-10 reached their maximum concentrations 1 hour after transplantation. Each increase in IL-6 correlated to a rise in IL-10. IL-2, IL-4, TNF-α, and IFN-γ had a particular time-course for each patient studied. Bilirubin fell to almost normal values but not in cases of postoperative complications, where IL-6 showed values four times higher compared to those of liver transplant recipients who did not show postoperative complications. IL-6 and IL-10 plasma concentrations and serum bilirubin level might be useful as a predictive factor of postoperative complications in liver transplant recipients.

Predictive factors comparison of complications and recurrences in three tension-free herniorraphy techniques

Inguinal herniorraphy is one of the most common surgeries performed. Avoiding hernia recurrence is a primary concern. Hence, it is necessary to analyze the predictive factors of postoperative complications and recurrence. To compare the predictive factors of postoperative complications and recurrences of hernias among three tension-free open herniorraphy techniques. Five hundred and fifty-one inguinal hernia patients, operated on with one of three tension-free anterior approach herniorraphy techniques, were included in a cohort study. The three techniques were: Lichtenstein (n=214), Mesh-Plug (n=201) and Prolene Hernia System (PHS, n=136). The patients were evaluated at 15 days, 1 month, 6 months and then every year up to 5 years after hernioplasty. The variables evaluated were recurrences and postoperative complications. Relative risk was estimated from a univariate analysis of the presumable risk values, after which a multivariate analysis was carried out. Complications [n=27, (4.9%)] were more frequently associated with incarcerated hernia, a coexisting disease at the time of operation, hospitalization longer than 1 day, previous herniorraphy, a herniary ring larger than 4.5 cm and a history of postoperative complications. Recurrence was greater for the Mesh-Plug group [n=5 (2.5%), RR: 4.35 (CI: 0.85-22.23)] than for the Lichtenstein [n=2 (0.9%), RR: 0.63 (0.06-3.87)] and PHS [n=0 (0%), RR de 0 (0-2.39)] groups. The presence of coexisting disease during hernioplasty, incarcerated hernia and an extended hospital stay are predictive factors for hernia complications. Previous herniorraphy, a herniary ring larger than 4.5 cm and postoperative complications are predictive factors for hernia recurrence.

Age as a Prognostic Factor for Complications of Major Head and Neck Surgery

To determine if age alone is a prognostic indicator of surgical outcomes for major head and neck procedures. Retrospective cohort study over a 4-year period. Academic referral center, institutional practice, hospitalized care. Included in this study were patients who had undergone ablative, reconstructive, and other major surgical procedures of the head and neck, including neck dissection, laryngectomy, maxillectomy, thyroidectomy with lymphadenectomy, and composite resection of the oral cavity with reconstruction, for both malignant and benign disease. Patient data and intraoperative and postoperative course factors were recorded. Comorbidity was graded using an Adult Comorbidity Evaluation 27 test, Charlson Comorbidity Index, and American Society of Anesthesiology score. Postoperative complications were dichotomized, and multiple logistic regression was used for data analysis. Medical chart review identified 157 cases. Analysis of data revealed that time under general anesthesia was the only factor consistently related to complications (P<.006), and it was the only factor consistently related to length of stay (P<.001). Analysis of major complications (6% incidence) as an outcome using univariate analysis resulted in a strong positive correlation with both comorbidity indexes: Adult Comorbidity Evaluation 27 (P = .002) and Charlson Comorbidity Index (P = .005). Multiple logistic regression showed no significant relationship between age 70 years or older (20% of patients) and either complications or hospital length of stay. Patient's age alone is not a prognostic indicator of surgical outcome for major head and neck procedures. However, comorbidity is an important predictive factor for postoperative complications in any age group. Time under general anesthesia showed a statistically significant relationship with complication rate and hospital length of stay in multivariate analyses. Consequently, prevention of complications should focus on optimizing preoperative comorbid conditions.

Immunonutrition risk factors of respiratory complications after esophagectomy

Background There are many reports suggesting predictive factors for respiratory complications after esophagectomy, but few studies have focused on this problem from the aspect of immunonutrition. Methods A series of 45 esophageal cancer patients who underwent potentially curative resection between April 1996 and March 2001 was enrolled in this study. Preoperative and intraoperative variables were analyzed for a correlation between risk factors and respiratory complications. In this study, immunonutrition variables as assessed by ultrasonography and indirect calorimetry were used. Uni- and multivariate analyses were performed to determine the predictive factors. Results Of 45 patients, nine patients (20%) developed respiratory complications. In univariate analysis, forced expiratory volume in 1 s per body surface area (m 2) in preoperative respiratory parameters, respiratory quotient, and caloric contributions of fat (percentage) and carbohydrate according to indirect calorimetry, and serum CH50 level significantly influenced the occurrence of postoperative respiratory complications. In multivariate analysis, the caloric contribution of fat (percentage) by indirect calorimetry and serum α 1-antitrypsin (serine protease inhibitor) independently affected the occurrence of respiratory complications. Conclusions Preoperative excessive fat oxidation (potential starvation) and α 1-antitrypsin were independent predictive factors for postoperative complications. Nutrition support such as enteral feeding to improve this malnourished state would reduce the incidence of postoperative respiratory complications.

Obstructive jaundice due to the lower bile duct lesions: biliary drainage and predictive factors of postoperative complications

Obstructive jaundice due to the lower bile duct lesions: biliary drainage and predictive factors of postoperative complications

Complications and risk factors after ileal pouch-anal anastomosis for ulcerative colitis associated with primary sclerosing cholangitis.

This study determined predictive factors for postoperative complications and outcome after ileal pouch-anal anastomosis in patients with ulcerative colitis and primary sclerosing cholangitis. Patients with ulcerative colitis and primary sclerosing cholangitis treated by colectomy and ileostomy are at high risk of troublesome bleeding from peristomal varices. Postoperative complications and outcome were assessed in 40 patients with ulcerative colitis and sclerosing cholangitis who received an ileal pouch-anal anastomosis between January 1981 and February 1990. Immediate postoperative and remote ileoanal anastomosis-related complications were high but related directly to the severity of liver disease. No patient had perianastomotic anal bleeding. In patients with both ulcerative colitis and primary sclerosing cholangitis, ileal pouch-anal anastomosis is safe and is not associated with perianastomotic bleeding.